1. Field of the Invention
The present invention relates to whether or not network coding is being operated in a network node and a method of applying network coding in order to minimize an increase in cost, delay, and complexity in accordance with an application of network coding
2. Discussion of the Related Art
Network coding refers to a technical method of compressing packets into one or more packets. Unlike a store-and-forward method, which corresponds to a transmission method of the conventional network node, the network coding method stores packets applied and introduced to the network node in a buffer. Then, until a transmission opportunity occurs in an output link, the network coding method compresses the packets, which are accumulated in the buffer, to one or more packets by using an encoding method. Thereafter, the network coding method transmits the one or more compressed packets.
A transmitting node of a network coding based network may or may not divide information that is to be transmitted to block units and combine the generated blocks by performing an encoding process. An internal network node combines packets, which are transmitted from the transmitting node and introduced to the internal network node, by performing an encoding process before transmitting the corresponding packets. Thereafter, the internal network node delivers the combined packets to a node connected to the internal network node. A final receiving node, which intends to receive content, receives a transmitted packet, which consists of a combination of multiple packets being combined by an encoding process. Accordingly, in order to recover information that is intended to be transmitted by the transmitting node, when sufficiently receiving a linearly independent packet, the receiving node may recover the initial (or original) information, which is transmitted by the transmitting node, by resolving a system of linear equations.
With respect to a flow of packets, which are encoded by a network intermediate node, the network coding method may be divided into Intra-session network coding and Inter-session network coding. Intra-session network coding corresponds to a method of encoding only packets belonging to the same flow, among a plurality of packets being flown into the network intermediate node, and transmitting the encoded packets. Herein, since the receiving node is more likely to receive linearly independent packets, Intra-session network coding is more advantageous in recovering the initial information. However, there are limitations in the Intra-session network coding method for achieving network-optimized transmission efficiency via network coding.
Inter-session network coding corresponds to a method of encoding packets belonging to different flows, among a plurality of packets being flown into the network intermediate node, and transmitting the encoded packets. Inter-session network coding corresponds to a coding method, which is applied to a network, wherein a plurality of flows is generated and transmitted in order to allow multiple transmitting nodes and receiving nodes to perform communication at the same time, such as the Internet, so as to resolve network congestion and to achieve network-optimized transmission efficiency via inter-flow network coding. However, the inter-session network coding method is disadvantageous in that the coding method is more complicated, as compared to the intra-session network coding method, and, since it is more difficult for the receiving node to sufficiently receive linearly independent packets for information recovery, the number of flows corresponding to encoding is limited.
Network coding technology may send out (or transmit) multiple network packets at the same time by combining the packets. Accordingly, by efficiently using network resources, the network coding technology may enhance network processing capacity, and, in bottleneck sections, the network coding technology may control network congestion. However, in order to apply network coding, a calculation procedure for performing the encoding operation between a transmitting node and a network intermediate node is required, which causes an increase in complexity. Additionally, the network coding technology also requires a storage space, which is configured to temporarily store packets, which correspond to encoding targets, and a calculating capability, which is required for recovering initial (or original) packets from a receiving node. Therefore, in order to apply the network coding technology to the current networks, the above-described disadvantages of the network coding technology are required to be resolved.
Intra-session network coding is advantageous in that the complexity level is relatively low and that the likelihood of recovering the initial content from the receiving node is high. However, intra-session network coding is disadvantageous in that the efficiency in using network resources is lower than the efficiency of the inter-session coding method. And, the inter-session coding method is theoretically advantageous in that it can achieve maximum transmission efficiency available in the network. However, the inter-session coding method is disadvantageous in that it has a high complexity level and that limitations are required to be made in the application environment in order to recover the initial content from the receiving node.